In 2,000 consecutive stroke patients collected in a prospective hospital-ba
sed stroke registry over a 10-year period, we assessed whether stroke in me
n and women was different in respect to vascular risk factors, clinical fea
tures and natural history. The frequency of the different variable in men a
nd women was analyzed by means of univariate analysis and logistic regressi
on models. Women accounted for 48% of the study population (n = 967) and we
re older than men (mean age 75 vs. 69 years, p < 0.001). In the age group o
f 85 years or older, stroke was more frequent in women than in men (69.8 vs
. 30.2 %, p < 0.001). Women showed a higher frequency of cardioembolic infa
rction and a lower occurrence of lacunar infarction and stroke of undetermi
ned cause than men. In-hospital mortality (17.4 vs. 13.3%) and length of ho
spital stay (19.6 vs. 16.7 days) was significantly higher (p < 0.001) in wo
men than in men. In the model based on demographic variables and cardiovasc
ular risk factors, obesity, heart failure, atrial fibrillation and age were
significant predictors of stroke in women, while intermittent claudication
, ischemic heart disease, chronic obstructive pulmonary disease, cigarette
smoking and alcohol abuse were predictors in male sex. Hypertension and lim
b weakness were predictors for stroke in women, and absence of neurological
deficit at hospital discharge, lacunar syndrome and ataxia were predictors
in men in the models based on all variables. Women differ from men in the
distribution of risk factors and stroke subtype, stroke severity and outcom
e. Differences in stroke pathology and/or differences in functional anatomy
or plasticity of the brain between sexes may account for these findings. C
opyright (C) 2001 S. Karger AG, Basel.